The mechanisms for stabilizing the tertiary structure of a soluble globular protein and a membrane protein are viewed to be inherently different. The difference arises from the absence of the hydrophobic effect, as a maintaining force in the tertiary structure of membrane proteins that are largely embedded in the hydrophobic core of the lipid bilayer. The studies proposed here follow from a simple hypothesis: Polar interactions control membrane protein stability. This hypothesis will be examined with two different approaches: structural studies on synthetic peptide analogs of membrane protein sequences, and structural studies on intact bacteriorhodopsin and peptide fragments derived from its enzymatic cleavage. Synthetic peptides, that mimic the structural features of one or two Alpha-helical segments of bacteriorhodopsin, will be studied in solution and reconstituted into phospholipid vesicles. The goal of these studies is to determine the relative contribution of hydrogen bonding and ion pairing to peptide-peptide or intra-peptide interactions resulting from the interaction of secondary structural domains. The techniques to be used in these studies are intrinsic fluorescence anisotropy, circular dichroism, high resolution 1H-NMR, differential scanning calorimetry, and chemical crosslinking. Studies on bacteriorhodopsin are directed towards (1) distinguishing which calorimetrically observed transitions are derived from protein-protein interactions and which are from intra-protein interactions; and (2) what forces predominate in each type of interaction. Purple membrane, reconstituted monomeric and oligomeric bacteriorhodopsin, and reconstituted fragments from enzymatically cleaved bacteriorhodopsin will be studied by differential scanning calorimetry and temperature dependent visible spectroscopy, circular dichroism and fluorescence spectroscopy under a variety of solution conditions. The long term goal of these studies is the semi-synthetic regeneration of bacteriorhodopsin structure from both synthetic and native components. This success will result from the accumulated understanding of the forces that contribute to the stability of a membrane protein.